ESS-MOTIFS: Discovering Rubric-Aligned Motifs for Cohort-Level Essay Assessment

Ngoc Thi Nguyen, N. Duane Loh

Submission Type: I want my submission to be considered for both oral and poster presentation.

Keywords: Essay Assessment, Motif Discovery, Interpretable Motifs, Retrieval-Augmented Generation (RAG), Large Language Models (LLMs)

TL;DR: ESS-MOTIFS is a rubric-aligned motif analysis framework that identifies prevalent patterns across essays and uses retrieval-augmented LLMs to generate synthesized summaries, enabling interpretable, auditable, and scalable cohort-level assessment.

Confirmation Of Submission Requirements: I submit an abstract. It uses the template provided on the submission page and is no longer than 2 pages.

Submission Number: 365